Natural Language Search and automated dialog or question and answer (“Q/A”) systems (sometimes referred to as chat bots) may be deployed to assist users on electronic commerce (E-Commerce) web sites. Natural Language Search applications provide a type of computer human interface in web sites wherein relevant linguistic items such as verbs, phrases and clauses are extracted from a user's natural language query and then used to select associated data within the site. A natural language search engine attempts to find targeted answers to user questions expressed or implied by the words within a given query. Rather than merely searching for keywords within the search query and returning results matching the individual keywords, natural language processing attempts to identify a query, topic or question defined by a plurality of query words taken as a whole, as a function of considering keywords in combination with each other or as modified by verbs, adverbs, adjectives, subject clauses, etc., associated therewith, or in context of services or products offered on the web site hosting the query, and then search and return results that likely contain answer to the “question” or topic recognized as expressed or implied by the plurality of words in the query.
An automated dialog system is a type of conversational agent, a computer program designed to simulate an intelligent conversation with one or more human users via auditory or textual methods to answer a user's questions and guide a user through a given web site. Such programs are often designed to simulate how a human would behave as a conversational partner. They generally have higher rates of success and customer satisfaction when the domain of possible topics is more limited, relative to a more unlimited, universal domain of possible topics, as the questions posed in response to each user input is thereby more likely to be on point to the information sought by the user. Generally automated dialog systems are deployed for customer service and information acquisition from a particular line of products or services on a host web site. While some automated dialog system use natural language processing systems, it is generally faster and more efficient for such systems to scan for relevant keywords within a user input and return results from the site databases that are ranked in relevancy as a function of number of matching keywords, degree of similarity to a query wording pattern, etc.